Machine learning-based seismic response and performance assessment of reinforced concrete buildings

Abstract Complexity and unpredictability nature of earthquakes makes them unique external loads that there is no formula used for the prediction seismic responses. Hence, this research aims to implement most well-known Machine Learning (ML) methods in Python software propose a model response performance assessment Reinforced Concrete Moment-Resisting Frames (RC MRFs). To prepare 92,400 data points training dataset developing data-driven techniques, Incremental Dynamic Analyses (IDAs) were performed considering 165 RC MRFs with two-, twelve-Story elevations having bay lengths 5.0 m, 6.1 7.6 m assuming near-fault excitations. Then, important structural features considered datasets train test ML-based models, which improved innovative techniques. The results show algorithms have higher R 2 values estimating Maximum Interstory Drift Ratio (IDR max ), two artificial neural networks extreme gradient boosting can estimate Median IDA curves (M-IDAs) MRFs, be limit-state capacity existing or newly constructed buildings. validate generality accuracy proposed model, five-Story building different input was used, are promising. Therefore, graphical user interface introduced as user-friendly tool help researchers buildings, while reducing computational cost analytical efforts.